CN-122027582-A - Power communication resource elastic allocation method and system based on multi-objective optimization

CN122027582ACN 122027582 ACN122027582 ACN 122027582ACN-122027582-A

Abstract

The invention belongs to the technical field of digital communication and network flow control, and particularly relates to a power communication resource elastic allocation method and system based on multi-objective optimization, wherein the method comprises the steps of extracting space distribution difference according to real-time flow state data and calculating a flow space-time evolution gradient by combining with historical flow rate surge state; the method comprises the steps of extracting physical cache boundary according to node attribute characteristics, fusing the physical cache boundary with traffic space-time evolution gradient to calculate information physical weakness, carrying out self-adaptive weighting processing according to the information physical weakness and service priority to generate an elastic allocation factor, allocating actual available bandwidth quota according to the elastic allocation factor, and refreshing rate configuration parameters. The method and the device can identify the key communication nodes which are in danger of congestion limit and perform physical bandwidth compensation, ensure that the key signaling obtains the priority forwarding resource, and improve the network communication robustness.

Inventors

CHEN MINGYUAN
LIU ZONGPU
SHI XIN
LIANG LI
MA LUTONG
Quan Zeying
ZHANG SHIMIN
LIU YANHUI
FANG JIANXIN

Assignees

信正聚力(陕西)计量检测有限公司

Dates

Publication Date: 20260512
Application Date: 20260410

Claims (10)

1. The power communication resource elastic allocation method based on multi-objective optimization is characterized by comprising the following steps: periodically collecting real-time traffic state data and node attribute characteristics of communication nodes in a global power communication network, wherein the real-time traffic state data comprises traffic rates of a target communication node and adjacent communication nodes thereof, and the node attribute characteristics comprise service priority, queue length and maximum cache capacity of the target communication node; extracting space distribution difference according to the flow rate of the target communication node and the flow rate of the adjacent communication node, and calculating the flow space-time evolution gradient of the target communication node by combining the historical flow rate surge state of the target communication node; Extracting a physical cache boundary according to the queue length and the maximum cache capacity of the target communication node, fusing the physical cache boundary with a traffic space-time evolution gradient, and calculating the information physical weakness of the target communication node; And distributing actual available bandwidth quota for the target communication node according to the elastic distribution factor, and refreshing rate configuration parameters of a bottom hardware forwarding plane of the target communication node based on the actual available bandwidth quota.
2. The power communication resource elastic allocation method based on multi-objective optimization according to claim 1, wherein the periodically collecting real-time traffic status data and node attribute characteristics of communication nodes in a global power communication network comprises: And reading the bottom hardware register data in each communication node management information base by using a simple network management protocol to acquire the real-time traffic state data and node attribute characteristics of each communication node in the current time window.
3. The power communication resource elastic allocation method based on multi-objective optimization according to claim 1, wherein the steps of extracting a spatial distribution difference according to a traffic rate of a target communication node and a traffic rate of an adjacent communication node, and calculating a traffic space-time evolution gradient of the target communication node in combination with a historical traffic rate surge state of the target communication node include: Calculating the space flow divergence of the target communication node according to the difference between the flow rate of the target communication node and the flow rate of the adjacent communication node, and calculating the flow space-time evolution gradient of the target communication node according to the flow rate change condition of the target communication node in each time window and the space flow divergence of the target communication node.
4. The multi-objective optimization-based power communication resource elastic allocation method according to claim 3, wherein the spatial traffic divergence satisfies an expression: ; Wherein, the Representing a target communication node In the current time window Spatial flow divergence of (2); representing a target communication node Is a total number of neighboring communication nodes; Representing local traversal sequence numbers of adjacent communication nodes in summation calculation; Representing a natural logarithmic function; representing a target communication node In the current time window Is a flow rate of (1); Represent the first The current time window of each adjacent communication node Is a flow rate of (1); Representing local traversal sequence numbers of adjacent communication nodes in average value calculation; Represent the first The current time window of each adjacent communication node Is provided).
5. The multi-objective optimization-based power communication resource elastic allocation method according to claim 3, wherein the flow space-time evolution gradient satisfies the expression: ; Wherein, the Representing a target communication node In the current time window A flow space-time evolution gradient of (2); An exponential function that is based on a natural constant; representing a target communication node In the current time window Is a flow rate of (1); representing the number of historical time windows; a traversal variable representing a historical time window; representing a target communication node In the first place The flow rates of the historical time windows; representing a target communication node In the current time window Spatial flow divergence of (2).
6. The power communication resource elastic allocation method based on multi-objective optimization according to claim 1, wherein the extracting the physical buffer boundary according to the queue length and the maximum buffer capacity of the objective communication node, and fusing the physical buffer boundary with the traffic space-time evolution gradient, and calculating the information physical vulnerability of the objective communication node comprises: And calculating the physical cache exhaustion trend of the target communication node according to the queue length of the target communication node and the maximum cache capacity of the target communication node, and calculating the information physical weakness of the target communication node according to the physical cache exhaustion trend of the target communication node and the traffic space-time evolution gradient of the target communication node.
7. The multi-objective optimization-based power communication resource elastic allocation method according to claim 6, wherein the physical buffer exhaustion trend satisfies an expression: ; Wherein, the Representing a target communication node In the current time window Is a physical cache exhaustion trend; An exponential function that is based on a natural constant; representing a target communication node In the current time window Is a queue length of (1); representing a target communication node Is used for the maximum buffer capacity of the system.
8. The multi-objective optimization-based power communication resource elastic allocation method according to claim 6, wherein the information physical weakness degree satisfies an expression: ; Wherein, the Representing a target communication node In the current time window Is the information physical weakness of (a); representing a target communication node In the current time window A flow space-time evolution gradient of (2); Representing local traversal sequence numbers of adjacent communication nodes in summation calculation; representing a target communication node Is a total number of neighboring communication nodes; Represent the first The current time window of each adjacent communication node A flow space-time evolution gradient of (2); representing a target communication node In the current time window Is a tendency for physical cache exhaustion.
9. The multi-objective optimization-based power communication resource elastic allocation method according to claim 1, wherein the elastic allocation factor satisfies an expression: ; Wherein, the Representing a target communication node In the current time window An elastic partitioning factor of (a); representing a target communication node In the current time window Is the information physical weakness of (a); An exponential function that is based on a natural constant; is a priority amplification factor; representing a target communication node Is a priority of service of (1); representing traversal sequence numbers of all communication nodes in the global communication network; representing a set of all communication nodes in the global communication network; representing the highest service priority of all communication nodes bearing services in the global communication network; A traversal sequence number representing a global communication node participating in resource scheduling; representing the total number of communication nodes participating in resource scheduling in the global communication network; Represent the first The current time window of each communication node Is the information physical weakness of (a); Represent the first Traffic priority of each communication node.
10. A multi-objective optimization based power communication resource elastic allocation system comprising a processor and a memory storing computer program instructions which, when executed by the processor, implement the multi-objective optimization based power communication resource elastic allocation method according to any one of claims 1-8.

Description

Power communication resource elastic allocation method and system based on multi-objective optimization Technical Field The invention relates to the technical field of digital communication and network flow control. More particularly, the invention relates to a power communication resource elastic allocation method and system based on multi-objective optimization. Background With the deep development of the smart power grid, the power communication network based on a specific industrial communication standard carries massive key services such as relay protection, a wide area measurement system, substation automation control and the like, and when extreme working conditions such as short circuit faults and the like occur, massive alarm messages instantaneously burst, so that extremely large transient bearing pressure can be brought to a forwarding framework of bottom communication hardware equipment. In the current power communication resource allocation field, a conventional bandwidth scheduling strategy generally adopts a static time division multiplexing technology or a weighted fair queuing mechanism based on a fixed priority, and the conventional queuing mechanism mainly relies on a fixed weight coefficient configured in advance by a system to carry out polling distribution of resources on each port. However, the existing scheduling mechanism lacks dynamic perceptibility of evolution trend of traffic on spatial network topology, deep mapping relation between port cache backlog of a switch and physical topological stress of a power grid cannot be deeply analyzed, so that the scheduling mechanism cannot adapt to nonlinear growth characteristics of burst traffic, and when transient burst large-traffic impact is faced, fixed weight polling often causes rapid accumulation of buffer queues of core aggregation nodes, so that the load limit is easily broken through and a tail discarding mechanism is triggered. The congestion failure of the communication link directly causes relay protection action tripping signaling with severe time delay constraint to be forcibly discarded due to queuing timeout, the loss of key signaling can further block the issuing of a load cutting instruction of a dispatching control central cutting machine, the best opportunity for isolating a fault source is delayed by the dispatching defect of the bottom layer, and the local fault is finally induced to be accelerated to evolve into a large-area power failure accident, so that the overall safe and stable operation of the novel power system is seriously threatened. Disclosure of Invention In order to solve the technical problem that the static scheduling cannot adapt to the burst traffic so as to discard the critical signaling, the invention provides schemes in the following aspects. In a first aspect, the present invention provides a power communication resource elastic allocation method based on multi-objective optimization, including: periodically collecting real-time traffic state data and node attribute characteristics of communication nodes in a global power communication network, wherein the real-time traffic state data comprises traffic rates of a target communication node and adjacent communication nodes thereof, and the node attribute characteristics comprise service priority, queue length and maximum cache capacity of the target communication node; extracting space distribution difference according to the flow rate of the target communication node and the flow rate of the adjacent communication node, and calculating the flow space-time evolution gradient of the target communication node by combining the historical flow rate surge state of the target communication node; Extracting a physical cache boundary according to the queue length and the maximum cache capacity of the target communication node, fusing the physical cache boundary with a traffic space-time evolution gradient, and calculating the information physical weakness of the target communication node; And distributing actual available bandwidth quota for the target communication node according to the elastic distribution factor, and refreshing rate configuration parameters of a bottom hardware forwarding plane of the target communication node based on the actual available bandwidth quota. Preferably, the periodically collecting real-time traffic status data and node attribute characteristics of communication nodes in the global power communication network includes: And reading the bottom hardware register data in each communication node management information base by using a simple network management protocol to acquire the real-time traffic state data and node attribute characteristics of each communication node in the current time window. Preferably, the extracting the spatial distribution difference according to the traffic rate of the target communication node and the traffic rates of the adjacent communication nodes, and calculating the traffic space-time evoluti